Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Nikon D5300 ADU when Dealing with flats

  • Please log in to reply
8 replies to this topic

#1 pscm

pscm

    Mariner 2

  • *****
  • topic starter
  • Posts: 292
  • Joined: 27 Jun 2009
  • Loc: Maine

Posted 21 August 2019 - 09:54 PM

Last night was having my first full image session using the asiair. When trying to do flats to get the ADU in the right range on the histogram on the asiair they were totally saturated.  I was doing some research and it would maybe seem that DSLR's may actually be be fully saturated at quite a bit less of a normal histogram.  I was trying to find a tool that could convert an NEF file to a fits as there are some programs out there to give a true measurement but struggling to find something to convert the files.  I am using star tools so don't believe their is a good histogram there.



#2 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 16821
  • Joined: 27 Oct 2014

Posted 21 August 2019 - 10:14 PM

http://www.astrosurf...s-software.html

 

"Load a RAW file" command loads NEF.  You can then save it as a FIT, or whatever.

 

You can also view the histogram of the NEF directly just fine.  You may need to use options to zoom in on the piece you want to examine closely.

 

I see no sign my D5500 saturates in anything but the to be expected manner.


Edited by bobzeq25, 21 August 2019 - 10:26 PM.

  • Michael Covington likes this

#3 Michael Covington

Michael Covington

    Author

  • *****
  • Vendors
  • Posts: 5522
  • Joined: 13 May 2014
  • Loc: Athens, Georgia, USA

Posted 21 August 2019 - 10:56 PM

Right, I have had no trouble taking good flats with my D5300 and D5500.   Judge them like daytime photos, viewing the image and the histogram on the camera's display.  Put them a little lighter than mid-histogram.



#4 pscm

pscm

    Mariner 2

  • *****
  • topic starter
  • Posts: 292
  • Joined: 27 Jun 2009
  • Loc: Maine

Posted 22 August 2019 - 11:31 AM

Thanks.  I will do some work.  It was strange with the asiair that if I tried to take a flat to get the histogram on the software I needed like 3 seconds and even then I couldn't get to mid level but then the pic was black.  Probably a software issue as I know it is still a work in motion.  Luckily I had not moved anything on my scope so I just took pics with the camera until I got them slightly above center.  this only took about 1/30th of a second with the light intensity I was using.  I probably should have lowered the light and hooked up my intervalometer but I was being lazy. and wanted to do some processing with my new ability to run 5 min subs. 



#5 pscm

pscm

    Mariner 2

  • *****
  • topic starter
  • Posts: 292
  • Joined: 27 Jun 2009
  • Loc: Maine

Posted 23 August 2019 - 06:00 AM

Guess I am not understanding how to use the program listed above.  I open a pic and under the view tab and select histogram.  I see this particular histogram has a max of 26500 on the x axis. Which is not the standard max adu.



#6 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 16821
  • Joined: 27 Oct 2014

Posted 23 August 2019 - 09:06 AM

Guess I am not understanding how to use the program listed above.  I open a pic and under the view tab and select histogram.  I see this particular histogram has a max of 26500 on the x axis. Which is not the standard max adu.

Yep.  You can change the x axis under options.  The 14 bit max is 16384.

 

Not that important for flats.  You need the peak or peaks somewhere near the middle. 

 

Getting this "exact" does not change your images.  I have seen some people recommend a little higher than middle for the D5XXX, as in #3 here.

 

The reason you do this is simply to stay away from the edges, which can be non-linear.  The usual problem, if there is one, is too dark corners.  There's considerable leeway.


Edited by bobzeq25, 23 August 2019 - 09:20 AM.

  • Michael Covington likes this

#7 pscm

pscm

    Mariner 2

  • *****
  • topic starter
  • Posts: 292
  • Joined: 27 Jun 2009
  • Loc: Maine

Posted 24 August 2019 - 07:52 AM

Out last night and I noticed the asiair histogram was at 32767.5 which is actually 1/2 of full ADU.  The camera is set at 12 bit.  A bit odd and maybe a software thing.  I suppose one could use some logic to compensate for this. 



#8 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 16821
  • Joined: 27 Oct 2014

Posted 24 August 2019 - 09:37 AM

Out last night and I noticed the asiair histogram was at 32767.5 which is actually 1/2 of full ADU.  The camera is set at 12 bit.  A bit odd and maybe a software thing.  I suppose one could use some logic to compensate for this. 

Load a light into IRIS.  You can use IRIS to draw a slice through a fairly bright star with your mouse.  You first check "View slice".  That will immediately show you (the slice will have a flat top) where IRIS thinks saturation is (in ADU) with your complete setup.  Now you have a reference point to figure "one-half" from.

 

Logic not necessary.  <smile>  I really like operational definitions in AP.

 

I really like IRIS.   Have "View slice" unchecked.  Draw a box around a star.  Right click for a menu.  Click "PSF".  You now know the shape of the star.


Edited by bobzeq25, 24 August 2019 - 09:40 AM.

  • Michael Covington likes this

#9 TelescopeGreg

TelescopeGreg

    Apollo

  • -----
  • Posts: 1022
  • Joined: 16 Jul 2018
  • Loc: Auburn, California, USA

Posted 24 August 2019 - 11:38 AM

Out last night and I noticed the asiair histogram was at 32767.5 which is actually 1/2 of full ADU.  The camera is set at 12 bit.  A bit odd and maybe a software thing.  I suppose one could use some logic to compensate for this. 

Yeah, something is a bit odd.  ADU values are integers.  No ".5" in an integer, so somebody is doing some floating point math on whatever the actual value is.  Also, any software engineer is automatically suspicious of anything displaying 32767.  It's a magic value (15 bits of all 1's), often a foreboding one.  May be nothing (i.e. just cosmetic), or not.  It would be good to get a second opinion from a different piece of software.


Edited by TelescopeGreg, 24 August 2019 - 11:40 AM.

  • Michael Covington likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics